Remember me
The underrepresentation of racial and ethnic minoritized populations in cancer clinical trials is well-established [-], particularly among Black/African American adults [-]. Despite federal initiatives and policies aimed at increasing cancer clinical trial enrollment and participation rates of underrepresented groups, rates have not improved among people from racial and ethnic minoritized groups, and in some cases, the rates have even declined []. Attributable to factors across multiple levels of influence [], the underrepresentation of Black/African American adults in cancer clinical trials means that drugs and interventions are developed, tested, and disseminated to populations not reflective of the broader US cancer population, perpetuating health inequities [].
For example, 1 study found that Black/African American adults comprised only 7.4% of all participants in US Food and Drug Administration clinical trials that led to new, approved cancer drugs from 2014 to 2018 []. The participation-to-prevalence ratio reflects the representation of Black/African American adults in the clinical trial population relative to the general cancer population, where a ratio of 1 means there is identical or equal representation between groups. Across cancer types, the estimated participation-to-prevalence ratio for Black/African American US adults was 0.31, indicating significant underrepresentation in clinical trials that result in Food and Drug Administration approvals for cancer drugs []. Importantly, Black/African American adults are also less likely to participate in trials of novel treatments and technologies, such as precision oncology [,]. These disproportionately low rates of clinical trial participation among racial and ethnic minorities result in limited understanding by medical professionals and the greater research community of how well new diagnostic technology, treatment options, and supportive care services are working for racial and ethnic minorities in comparison to the predominantly White clinical trial participant population [,].
In addition to underrepresentation in cancer clinical trials, inequities in cancer care and survival rates persist [-]. Greater inclusion of Black/African American patients in cancer clinical trials is, therefore, essential to design and test interventions to address inequities in cancer care among Black/African American patients. For example, non-Hispanic Black/African American patients have significantly greater cancer diagnosis delay [], treatment delay [], and likelihood of diagnosis at an advanced cancer stage [] compared with non-Hispanic White patients. Even after accounting for cancer stage, cancer type, and other relevant covariates, Black/African American patients still have significantly lower survival rates than White patients [].
Prior studies have found that non-Hispanic, Black/African American patients have less awareness of cancer clinical trials and hold specific attitudes and beliefs about trial participation relative to non-Hispanic, White patients [,]. For example, in a qualitative study of Black/African American cancer survivors who received cancer treatment at a safety-net hospital, the primary clinical trial participation barriers were (1) limited knowledge and understanding of cancer clinical trials and (2) medical mistrust, fears, and other negative perceptions of cancer clinical trials. Participants also described wanting a peer (cancer survivor of a concordant race or ethnicity group) patient navigator who was well-versed in clinical trials knowledge and who could provide other forms of social support (eg, social or emotional, faith-based or spiritual, and instrumental support) []. These results were consistent with other studies emphasizing the roles of knowledge or awareness, medical mistrust, and social support in clinical trial enrollment; study participation; and retention over time [-].
Other specific attitudes held by Black/African American patients with cancer more than White patients include lower perceived cancer susceptibility and greater doubt about the usefulness and feasibility of translating cancer clinical trial results into clinical practice []. Other patient-level factors associated with less knowledge and awareness of cancer clinical trials include living in a rural area [], living farther away from universities or large hospital networks [], older age [], limited English language proficiency [], lower educational attainment [], and less annual household income []. Conversely, greater cancer clinical trial knowledge and the likelihood of trial participation are associated with a prior cancer diagnosis [], having a routine source of health care (ie, primary care access) [], and higher educational attainment []. Trial populations’ clinical knowledge and awareness are essential constructs for researchers to be aware of because the quality of communication between clinical trial staff and prospective trial participants is, in part, dependent upon patients’ clinical trial knowledge and confidence [].
Negative attitudes toward cancer clinical trials, particularly having greater concerns, are associated with cancer fatalism []. Other concerns cited by Black/African American patients with cancer associated with decreased cancer clinical trial intentions are greater fear of the unknown [], fear of death [], prior negative health care or clinical trial experiences [,,], fear of receiving an inferior treatment or placebo [], lower health literacy [,], anticipated discrimination [], and medical mistrust [,]. Structural racism, historical injustices, and unethical research practices have disproportionately affected Black/African American people and have perpetuated concerns of anticipated mistreatment by research personnel and broader medical mistrust [-]. However, levels of cancer-related knowledge and specific attitudes toward cancer clinical trials are associated with cancer clinical trial participation rates among Black/African American patients with cancer. For example, a qualitative study among Black men found that perceptions of greater research integrity and transparency were positively associated with willingness to participate in prostate cancer surveillance screening and clinical trials []. Other factors positively associated with willingness to participate in cancer research were having a family history of cancer, seeing greater value in screening and cancer prevention, and having more interest in learning about cancer and other health-related information [].
At the interpersonal level, Black/African American patients with cancer have differential access to cancer clinical trial information attributable to provider biases and patient-provider communication quality. For example, clinical trials are often initially discussed with patients by their health care providers, but provider bias, including racism and discrimination, results in less information sharing and discussion about cancer screenings, clinical trials, and cancer treatment options for Black/African American patients than for White patients []. At the clinic level, limited hiring of providers with language fluency beyond English reduces clinic access and decreases the feasibility of within-session information sharing about clinical trials for patients and families with limited English language proficiency []. Importantly, many Black/African American patients report not being offered a trial during their cancer care [-], despite overall positive perceptions of clinical trials, further exacerbating the inequity [].
Finally, it should be noted that individual-level awareness of clinical trials is only minimally helpful as an interventional target when structural and systemic factors more strongly drive participation rates. For example, studies have repeatedly demonstrated that some of the greatest barriers to clinical trial enrollment are inequitable clinical trial referrals and enrollment practices [] and stringent trial eligibility criteria [-]. Recent programs and initiatives implemented to increase awareness of cancer clinical trials among Black/African American patients have recognized that awareness must be addressed at multiple levels of influence to advance health equity. For example, a June 2022 article published by the American Society of Clinical Oncology suggests that clinics and health care facilities use 1 of 2 standardized clinic self-assessment tools to review their enrollment practices and patient-, provider-, and system-level barriers to clinical trial enrollment [-].
This study is a cross-sectional analysis of the baseline data from a parent randomized controlled trial (RCT) designed to evaluate the impact of a multicultural, clinical trial preparatory digital health tool (mychoice) or standard National Cancer Institute information for patients with cancer. mychoice was conceptualized and developed by a team of investigators at Fox Chase Cancer Center and the Temple University College of Public Health through extensive formative research with Black/African American patients, expertise in health disparities and clinical trial participation, commercial marketing techniques (perceptual mapping and vector message modeling), and best practices in digital health and patient engagement [,]. Although founded on clinical trial participation barriers significant to underrepresented patients, the tool is designed to be appropriate for all patients with cancer and to represent diverse patient perspectives.
ObjectivesA diverse sample of patients enrolled in the parent RCT completed a baseline survey before viewing the decision-making tool, providing an opportunity to explore racial disparities in a variety of factors previously linked to clinical trial participation rates and the clinical trial participation decision-making process. On the basis of the conducted formative work with Black/African American patients to inform the digital health tool used in the parent RCT, this study sought to confirm whether factors identified in the formative work were, in fact, salient to Black/African American patients with cancer relative to non–Black/African American patients with cancer at baseline. Findings will help explain Black/African American versus non–Black/African American participant responses to the culturally tailored, clinical trial decision-making tool and also help identify factors that could help further refine the decision-making tool. In addition, findings can be used to tailor and prioritize topics in provider education and training to better support the needs of Black/African American patients with cancer in cancer clinical trial decision-making.
The analytical sample at baseline included patients with cancer from 4 leading cancer centers in Philadelphia (Fox Chase Cancer Center, Temple University Hospital, University of Pennsylvania’s Abramson Cancer Center, and Thomas Jefferson University’s Sidney Kimmel Cancer Center) who consented to participate in the parent RCT (NCT03427177) and completed the baseline survey. Moreover, 3 of the 4 recruitment sites are National Cancer Institute–designated cancer centers. Eligible patients were actively being treated for cancer or in follow-up care (ie, within 6 months of definitive treatment), aged ≥18 years, able to speak and read English, and had not participated in a therapeutic clinical trial. The parent RCT had been planned to enroll 270 participants. In total, 257 participants consented and 249 (96.9%) completed the baseline survey. Patients of all racial and ethnic groups were eligible for the RCT, but only 244 (98%) of the 249 completed baselines reported valid or nonmissing data for their race and were analyzed in this study.
InstrumentsOverviewThe survey was developed using both validated instruments and study-related measures from formative work, including both qualitative interviews and surveys with Black/African American patients with cancer [,-]. Variables included in the present analyses were sociodemographic characteristics (ie, age, race, ethnicity, gender, income, educational attainment, insurance type, and cohabitation status), dichotomized race group (Black/African American vs non–Black/African American), clinical characteristics (ie, cancer stage and treatment status), general clinical trial knowledge, health literacy, cancer clinical trial perceptions (awareness, benefits, concerns, and cancer and health care experiences beliefs about health care providers and health), patient activation in cancer care, patient self-advocacy, self-efficacy in health care interactions, decisional conflict, and clinical trial intentions.
General Knowledge of Clinical TrialsGeneral knowledge of clinical trials was assessed using 16 revised items from Knowledge of Clinical Trials scale by Campbell et al []. Response options were “true” or “false” and were scored for accuracy. Scores were generated using the percentage of questions answered correctly, ranging from 0% to 100%.
Health LiteracyHealth literacy was assessed with a single item from the Single Item Literacy Screener, which specifically identifies adults who may need assistance reading and understanding health materials []. The item says, “How often do you need to have someone help you when you read instructions, pamphlets, or other written material from your doctor or pharmacy?” Response options were rated on a 5-point Likert scale, ranging from a score of “1” reflecting “never” to “5” reflecting “always.” On the basis of psychometric testing, scores >“2” reflect people with limited health literacy in reading and comprehending written health information [].
Cancer Clinical Trial PerceptionsPerceptions of cancer clinical trials were evaluated using 48 items developed by the primary investigators through formative work, reflecting domains of (1) awareness, (2) benefits, (3) concerns, (4) cancer and health care experiences, and (5) beliefs about health care providers and health [,-]. Response options were rated on an 11-point Likert scale ranging from 0 to 10 where “0” indicated strong disagreement and “10” indicated strong agreement. Item-level analyses were conducted in this study.
Patient Activation in Cancer CarePatient activation for cancer care decision-making was measured with 10-item Decisional Engagement Scale []. This instrument was developed specifically to understand patients’ level of involvement in their cancer care and engagement with active decision-making processes around treatment and care options []. Response options were rated on an 11-point Likert scale, ranging from 0 to 10 where “0” meant “doesn’t describe you at all” and “10” meant “perfectly describes you.” In psychometric evaluation, the 10-item Decisional Engagement Scale has demonstrated strong factor structure, reliability, and concurrent validity with health-related quality of life, shared decision-making preferences, and clarity about cancer care preferences [].
Patient Self-AdvocacyPatient self-advocacy was measured with 12-item Patient Self-Advocacy Scale []. Response options were rated on an 11-point Likert scale, ranging from 0 to 10. In addition, 1 item (“I don’t get what I need from my physician because I am not assertive enough”) was reverse coded before calculating an average summary score. The scale has demonstrated good internal consistency, construct validity, and criterion validity [].
Health Care Self-EfficacySelf-efficacy to engage with health care providers was measured with 10-item Perceived Self-Efficacy in Patient-Physical Interactions scale []. Items asked about confidence to do specific health care–related tasks, such as confidence to get a physician to listen to them, confidence in ability to know what questions to ask a physician, and confidence in ability to get a physician to take their health concerns seriously. Response options ranged from 1 to 5, where “1” indicated least confidence and “5” indicated most confidence [].
Decisional ConflictDecisional conflict about clinical trial participation was measured with 13-item Decisional Conflict scale proposed by O’Connor []. Response options were rated on a 5-point Likert scale, ranging from 0 to 4 where “0” reflected “strongly agree” and “4” reflected “strongly disagree.” Scoring of 4 subscales (uncertainty, informed, value clarity, and decision support) was done by summing the items within the subscale, dividing by the number of items within that subscale, and multiplying by 25. This resulted in a score ranging from 0 to 100. A total score for all items was also calculated by summing all items, dividing by 13, and multiplying by 25. This, too, led to a total score ranging from 0 to 100. In psychometric testing, the scale had good discriminant validity between those who choose versus those who do not choose to engage in a health behavior. Other psychometric properties were determined to be acceptable [].
Clinical Trial Participation IntentionsIntentions to participate in a cancer clinical trial were assessed with a single, modified item from the Choice Predisposition Scale proposed by O’Connor []. The item read, “We would like to know what your opinion is about your cancer treatment options at present. When your doctor asks you to make a choice about treatment methods, please indicate how strongly you agree or disagree that you would choose to participate in a clinical trial, if offered.” Response options ranged from 0 to 10, where “0” indicated strongly disagree,” a “5” meant “neither agree nor disagree,” and “10” indicated “strongly agree.” This scale has good psychometric properties, such as high test-retest validity, good construct validity, high sensitivity to change, and discriminant validity [].
ProceduresProspective participants were screened for eligibility (aged ≥18 years, cancer diagnosis, receiving current or follow-up care, English speaking, and had not previously participated in a clinical trial). Participants provided verbal informed consent either in person or over the phone. Consent was verified via an e-consent using REDCap (Research Electronic Data Capture; Vanderbilt University), a web-based application developed to capture data for research [,]. Consented patients were randomized to intervention conditions via REDCap and completed a baseline survey prior to viewing any intervention content. The baseline survey assessments were web-based and were conducted through REDCap. Patients could either complete the study at the hospital using a study iPad (Apple Inc) or at home on their own devices. The baseline survey took approximately 45 minutes.
Statistical AnalysisUnivariate statistics using means, SDs, and percentages are presented to characterize the participant sample. Differences in sociodemographic and clinical characteristics between dichotomous race groups (ie, Black/African American and non–Black/African American patients) were evaluated using chi-square tests of independence and independent sample 2-tailed t tests, as appropriate. Independent sample t tests were also used to examine for differences between Black/African American and non–Black/African American patients’ clinical trial knowledge, attitudes toward cancer clinical trials, and intentions to participate in a clinical trial. While some variables (eg, health literacy and self-efficacy in health care interactions) were highly skewed, t tests were still used as opposed to nonparametric testing because t tests are robust to skewed distributions when the sample size is >200 []. Homogeneity of variances between groups was evaluated for each item before running independent samples t tests, and the appropriate t test assumptions were applied accordingly. All data analyses were conducted in StataSE (version 17.0; StataCorp).
Ethical ConsiderationsThe study protocol was approved by the Fox Chase Cancer Center’s institutional review board (#17-8013). All procedures involving human participants were in accordance with the ethical standards of the institutional or national research committee and with the 1964 Declaration of Helsinki and its later amendments or comparable ethical standards. All participants provided verbal informed consent. Verification of consent with e-consent and all other study data were collected in REDCap, a secure web-based application developed to collect and store research data [,]. To protect participants’ privacy, the data were coded before analysis using unique participant study identifiers and no direct identifiers were in the analytic data set. Participants were compensated US $25 for completing the baseline survey, educational intervention, and the posttest survey. However, this paper describes results from the baseline survey data only.
compares sociodemographic and clinical characteristics by dichotomous race group. - show results of all remaining independent sample t tests for differences in average general clinical trials knowledge, health literacy, perceptions of cancer clinical trials, patient activation, patient advocacy, health care self-efficacy, decisional conflict, and clinical trial intentions by race group.
Table 1. Sociodemographic and clinical characteristics of study participants by race at baseline (N=244).CharacteristicsTotala (N=244)Non–Black/African American (n=149)Black/African American (n=95)Chi-square (df) or t test (df)P valueGender, n (%)7.465 (1).006aPercentages are rounded and, therefore, may not add up to 100%. Missing or invalid data were excluded from this table for insurance type (n=4), ethnicity (n=32), income (n=19), cohabitation status (n=4), cancer stage (n=51), and treatment status (n=18).
bGED: General Educational Development.
cNot applicable. This was because the table is split by binary race, so examining race by race is nonsensical.
dFisher exact test was used when one or more of the expected cell counts was <5.
eReceiving treatment includes treatment types, such as chemotherapy, radiation, surgery, and any other types of cancer treatment. Follow-up care includes posttreatment care within 6 months of the last receipt of treatment.
Table 2. Baseline knowledge, health literacy, awareness of clinical trials for Black/African American versus non–Black/African American oncology patients (N=244).aClinical trials knowledge was a percentage ranging from 0 to 100.
bHealth literacy ranged from 0 to 4, where higher values reflected lower health literacy.
cVariances were not equal between groups, so an independent sample t test with unequal variances was used.
dResponse options for awareness items ranged from 0 to 10, where 0 indicated strong disagreement and 10 indicated strong agreement.
Table 3. Baseline perceived benefits and concerns about cancer clinical trials for Black/African American versus non–Black/African American oncology patients (N=244).aResponse options for perception items ranged from 0 to 10, where 0 indicated strong disagreement and 10 indicated strong agreement.
bVariances were not equal between groups, so independent sample t test with unequal variances was used.
Table 4. Baseline health care experiences, health care beliefs, patient self-advocacy, patient activation, health care self-efficacy, decisional conflict, and intentions to participate in cancer clinical trials for Black/African American versus non–Black/African American oncology patients (N=244).aResponse options for perception items, patient activation, patient self-advocacy, and clinical trial intentions ranged from 0 to 10, where “0” indicated strong disagreement and “10” indicated strong
Comments (0)